Method for improving sensorless control performance of permanent magnet synchronous linear motor

By improving the second-order generalized integrator and normalization processing, the problems of magnetic flux drift and high-order harmonics in sensorless control of permanent magnet synchronous linear motors are solved, and the estimation accuracy of the rotor position and speed are improved, and the harshness of the Efficient control under working conditions.

CN114696695BActive Publication Date: 2024-11-29XIAN UNIV OF TECH
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202210004491.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-01-04
Publication Date
2024-11-29
Estimated Expiration
2042-01-04

AI Technical Summary

Technical Problem

In the existing permanent magnet synchronous linear motors without sensor control, the magnetic flux observer is susceptible to magnetic flux drift and high-order harmonic interference, affecting the accuracy of the rotor position estimation.

Method used

Using the permanent magnet synchronous linear motor magnetic resonance observer sensorless vector control system based on improved second-order generalized integrators, the improved second-order generalized integrator model is built and a new Back EMF is used to process the rotor magnetic resonance using a normalized method, configure bandwidth gain and frequency parameters, and realize dual closed-loop control of current and speed.

Benefits of technology

The accuracy of magnetic flux estimation is improved, the magnetic flux drift and high-order harmonic interference is reduced, and the accuracy of rotor position and speed estimation is enhanced, especially in the three-phase imbalance of the power grid.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114696695B_ABST
    Figure CN114696695B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for improving the sensorless control performance of a permanent magnet synchronous linear motor, which is specifically implemented according to the following steps: Step 1, obtaining the state equation of the permanent magnet synchronous linear motor after being described by establishing the state equation of the permanent magnet synchronous linear motor; Step 2, building an improved second-order generalized integrator model according to the state equation of the permanent magnet synchronous linear motor, and defining a new back electromotive force output by the second-order generalized integrator; Step 3, replacing the pure integrator with an improved second-order generalized integrator plus an integrator in the flux observer, establishing a sensorless control system of the permanent magnet synchronous linear motor flux observer based on the improved second-order generalized integrator, and using the "normalization" method to process the rotor flux to solve the jitter problem in the initial estimation of the electrical angular velocity that may be caused by the setting of the initial value of the integral. The present invention solves the problems of flux drift and high-order harmonics in the flux observer in the prior art.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of position sensorless control of permanent magnet synchronous linear motors, and in particular relates to a method for improving sensorless control performance of permanent magnet synchronous linear motors. Background Art

[0002] In recent years, linear motors have been widely used in industrial production and other fields. Rotary motors achieve reciprocating linear motion by connecting roller screws. However, permanent magnet linear motors do not require intermediate transmission devices and can directly achieve linear motion. The omission of intermediate transmission devices simplifies the entire control system, greatly improving its performance in terms of thrust, speed, accuracy, etc. Linear motors generally use linear gratings to detect the speed and position of the mover. Since linear motors currently need to operate in more occasions and even in harsher working conditions, the use of speed sensors reduces the robustness of permanent magnet linear motors and limits their application range, while also increasing installation costs.

[0003] Speed ​​sensorless control is not only an important research direction in modern AC transmission control technology, but also a key technology for studying high-performance direct drive control. So far, there are many feasible methods for sensorless control of permanent magnet linear motors, including Luenberger observer method, high-frequency injection method, Kalman filter method, flux observer, etc. Among them, the flux observer method has the advantages of small computational complexity and easy implementation, so it is a research hotspot.

[0004] The flux observer is an extension of the back-EMF observation method. It also has the characteristics of small calculation and easy implementation, so it has a wide range of application prospects. However, the estimation result of the mover position of the linear motor is easily affected by the flux drift and the interference of high-order harmonics caused by the nonlinearity of the inverter and the end effect. Therefore, how to improve the accuracy of the mover position estimation when the linear motor is running is the focus and difficulty of the research. Summary of the invention

[0005] The purpose of the present invention is to provide a method for improving the sensorless control performance of a permanent magnet linear synchronous motor, which solves the problems of flux drift and high-order harmonics in the flux observer in the prior art.

[0006] The technical solution adopted by the present invention is a method for improving the sensorless control performance of a permanent magnet synchronous linear motor, which adopts a permanent magnet synchronous linear motor flux observer sensorless vector control system based on an improved second-order generalized integrator for control, and the steps are as follows:

[0007] Step 1, in the α-β coordinate system, a state equation of the permanent magnet synchronous linear motor is obtained, and then the state equation of the permanent magnet synchronous linear motor is described using a state space to obtain the described state equation of the permanent magnet synchronous linear motor;

[0008] Step 2, according to the permanent magnet linear synchronous motor state equation, build an improved second-order generalized integrator model, and define a new back electromotive force output by the second-order generalized integrator;

[0009] Step 3, use the improved second-order generalized integrator plus integrator to construct an improved flux observer, and further configure the bandwidth gain, frequency and other parameters of the second-order generalized integrator by analyzing the transfer function; and use the "normalization" method to process the rotor flux to solve the jitter problem in the initial estimation of the electrical angular velocity due to the setting of the initial value of the integral.

[0010] The present invention is also characterized in that:

[0011] A sensorless vector control system for permanent magnet linear synchronous motor flux observer based on improved second-order generalized integrator d = 0, realizing the dual closed-loop control of current and speed; the three-phase current i a 、i b 、i c After being measured by the Hall sensor, the current i in the two-phase stationary coordinate system is first obtained through a 3s / 2s transformation. α 、i β , and then through 2s / 2r transformation, the two-phase rotating coordinate current i is obtained d 、i q The speed estimate value estimated by the flux observer is subtracted from the given speed value to obtain an error signal, which is then passed through the PI regulator to obtain the q-axis PI regulator current given value, and then the error signal obtained by subtracting it from the actual q-axis current value fed back is converted into the q-axis voltage after passing through the PI regulator. The error signal obtained by subtracting the given d-axis current from the actual value of the d-axis current is converted into the d-axis voltage after passing through the PI regulator. The voltage vector in the dq coordinate system can be obtained, and the voltage vector is then transformed by 2r / 2s to obtain the stationary coordinate system voltage vector u α 、u β Finally, six PWM signals are output through space vector pulse width modulation to drive the motor to move.

[0012] The mathematical model of the permanent magnet synchronous linear motor in step 1 is expressed as:

[0013]

[0014] In formula (1), u α 、u β are the components of the stator voltage on the α and β axes respectively; R is the stator resistance; ψ sα , sβ are the stator flux components of α and β axes respectively; i α 、i βare the components of the stator current on the α and β axes respectively; L s Set as the stator inductance of the permanent magnet linear synchronous motor;

[0015] The components of the rotor flux on the α and β axes can be expressed as:

[0016]

[0017] The relationship between the components of the rotor magnetic flux on the α and β axes and the electrical angle can be expressed as:

[0018]

[0019] In formula (3), ψ f is the permanent magnet flux of the permanent magnet linear motor;

[0020]

[0021] In formula (4), θ e is the position of the mover

[0022] The relationship between the mover position and the initial position can be expressed as:

[0023]

[0024] In formula (5), θ e is the position of the mover, τ is the pole pitch, and x is the initial position;

[0025] Therefore, the electrical angular velocity can be expressed as:

[0026]

[0027] In formula (6), ω e is the electrical angular velocity, v is the running speed of the linear motor;

[0028] The calculation of the mover flux is based on the following formula:

[0029]

[0030] At this time, the back electromotive force is expressed as follows:

[0031]

[0032] Step 2 includes:

[0033] Step 2.1, establish the LPF transfer function formula:

[0034]

[0035] qe′ is a 50Hz sinusoidal signal, so the cutoff frequency of LPF can be selected at 50Hz;

[0036] Step 2.2, the transfer function of the improved second-order generalized integrator is:

[0037]

[0038]

[0039] Where the output of D1(s) is e without DC component αβ ', the output of Q1(s) does not contain DC components and high-order harmonics and is consistent with e αβ 'Orthogonal qe αβ ';

[0040] For the selection of k value, the filtering effect and response speed of SOGI are taken into consideration. Generally, k is set between Nearby, it also meets the filter quality factor Q D2 =0.707; In order to achieve better dynamic performance, the ω′ in the improved second-order generalized integrator adopts the closed-loop feedback value of the estimated linear motor operating frequency;

[0041] Step 2.3, the back electromotive force output by the second-order generalized integral must be extracted through the positive sequence component as shown in formula (12):

[0042]

[0043] The permanent magnet linear synchronous motor flux observer sensorless vector control system based on improved second-order generalized integrator in step 3 mainly includes a speed loop, a current loop and an improved flux observer; the permanent magnet linear synchronous motor flux observer sensorless vector control system based on improved second-order generalized integrator adopts i d = 0, realizing the dual closed-loop control of current and speed; the three-phase current i a 、i b 、i c After being measured by the Hall sensor, the current i in the two-phase stationary coordinate system is first obtained through a 3s / 2s transformation. α 、i β , and then through 2s / 2r transformation, the two-phase rotating coordinate current i is obtained d 、i qThe speed estimate value estimated by the flux observer is subtracted from the given speed value to obtain an error signal, which is then passed through the PI regulator to obtain the q-axis PI regulator current given value, and then the error signal obtained by subtracting it from the actual q-axis current value fed back is converted into the q-axis voltage after passing through the PI regulator. The error signal obtained by subtracting the given d-axis current from the actual value of the d-axis current is converted into the d-axis voltage after passing through the PI regulator. The voltage vector in the dq coordinate system can be obtained, and the voltage vector is then transformed by 2r / 2s to obtain the stationary coordinate system voltage vector u α 、u β Finally, six PWM signals are output through space vector pulse width modulation to drive the motor to move.

[0044] The specific steps of step 3 are:

[0045] Step 3.1, the back electromotive force obtained by the second-order generalized integrator is passed through the integrator to obtain the stator flux:

[0046]

[0047] Step 3.2, according to the relationship between the rotor flux and the stator flux, we can get equation (14):

[0048]

[0049] Step 3.3, normalizing the rotor flux can obtain equation (15):

[0050]

[0051] Step 3.4, estimate the position of the mover, the electrical angular velocity of the mover and the speed of the mover by using the obtained mover flux as shown in equations (16)(17)(18):

[0052]

[0053]

[0054]

[0055] The beneficial effects of the present invention are:

[0056] The present invention improves the estimation accuracy of the flux linkage in the sensorless control system of the permanent magnet synchronous linear flux linkage observer, thereby improving the estimation accuracy of the position and speed of the mover. An improved second-order generalized integrator is added in front of the pure integrator to solve the problem of flux linkage drift in the flux linkage observer and filter out the problem of high-order harmonics caused by the nonlinearity of the inverter and the inherent end effect of the permanent magnet linear motor. In addition, when the system operates under the special working condition that the three-phase unbalanced grid voltage appears, the extraction of the positive sequence component ensures that the observation result of the flux linkage observer will not be distorted. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a system block diagram of a sensorless vector control system of a permanent magnet linear synchronous motor flux observer based on an improved second-order generalized integrator in step 1 of the present invention;

[0058] Figure 2 is a structural block diagram of the improved second-order generalized integrator in step 2 of the present invention;

[0059] Figure 3 It is a block diagram of the improved flux observer in step 3 of the present invention. DETAILED DESCRIPTION

[0060] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0061] The present invention is a method for improving the sensorless control performance of a permanent magnet synchronous linear motor. Figure 1 As shown, a permanent magnet linear synchronous motor flux observer sensorless vector control system based on an improved second-order generalized integrator is used for control, which mainly includes three parts: speed loop, current loop and improved flux observer.

[0062] A sensorless vector control system for permanent magnet linear synchronous motor flux observer based on improved second-order generalized integrator d = 0, realizing the dual closed-loop control of current and speed. a 、i b 、i c After being measured by the Hall sensor, the current i in the two-phase stationary coordinate system is first obtained through a 3s / 2s transformation. α 、i β , and then through 2s / 2r transformation, the two-phase rotating coordinate current i is obtained d 、i qThe speed estimate value estimated by the flux observer is subtracted from the given speed value to obtain an error signal, which is then passed through the PI regulator to obtain the q-axis PI regulator current given value, and then the error signal obtained by subtracting it from the actual q-axis current value fed back is converted into the q-axis voltage after passing through the PI regulator. The error signal obtained by subtracting the given d-axis current from the actual d-axis current is converted into the d-axis voltage after passing through the PI regulator, and the voltage vector in the dq coordinate system can be obtained. The voltage vector is then transformed by 2r / 2s to obtain the voltage vector u in the stationary coordinate system. α 、u β Finally, six PWM signals are outputted through space vector pulse width modulation to drive the motor to move. The present invention mainly includes the following steps:

[0063] Step 1, in the α-β coordinate system, a state equation of the permanent magnet synchronous linear motor is obtained, and then the state equation of the permanent magnet synchronous linear motor is described using a state space to obtain the described state equation of the permanent magnet synchronous linear motor;

[0064] Step 2, such as Figure 2 , according to the permanent magnet linear synchronous motor state equation, build an improved second-order generalized integrator model, and define a new back electromotive force output by the second-order generalized integrator;

[0065] Step 3, such as Figure 3 , an improved flux observer is constructed by using an improved second-order generalized integrator plus an integrator, and the bandwidth gain, frequency and other parameters of the second-order generalized integrator are further configured by analyzing the transfer function. And the "normalization" method is used to process the rotor flux to solve the jitter problem in the initial estimation of the electrical angular velocity that may be caused by the setting of the initial value of the integral.

[0066] In step 1, the mathematical model of the permanent magnet linear synchronous motor is expressed as:

[0067]

[0068] In formula (1), u α 、u β are the components of the stator voltage on the α and β axes respectively; R is the stator resistance; ψ sα , sβ are the stator flux components of α and β axes respectively; i α 、i β are the components of the stator current on the α and β axes respectively; L s Set as the stator inductance of the permanent magnet linear synchronous motor;

[0069] The components of the rotor flux on the α and β axes can be expressed as:

[0070]

[0071] The relationship between the components of the rotor magnetic flux on the α and β axes and the electrical angle can be expressed as:

[0072]

[0073] In formula (3), ψ f is the permanent magnet flux of the permanent magnet linear motor.

[0074]

[0075] In formula (4), θ e is the position of the mover

[0076] The relationship between the mover position and the initial position can be expressed as:

[0077]

[0078] In formula (5), θ e is the position of the mover, τ is the pole pitch, and x is the initial position.

[0079] Therefore, the electrical angular velocity can be expressed as:

[0080]

[0081] In formula (6), ω e is the electrical angular velocity, and v is the running speed of the linear motor.

[0082] The calculation of the mover flux is based on the following formula:

[0083]

[0084] At this time, the back electromotive force is expressed as follows:

[0085]

[0086] The specific measures in step 2 are:

[0087] Step 2.1, the two outputs of the improved second-order generalized integrator are e and αβ ' and with e αβ 'Orthogonal but without DC component and high-order harmonics qe αβ '. So qe αβ' can avoid the influence of DC bias caused by pure integral joints and can also avoid the problem of high-order harmonics caused by factors such as inverter nonlinearity and end effects. Compared with the traditional SOGI, the improved SOGI adds a low-pass filter (LPF) to the output error feedback controller kε and qv” channel, and outputs qe’ after subtracting it from qe”. The purpose of adding a low-pass filter is to make qe’ have a greater attenuation in the high frequency band, and the LPF transfer function is established as:

[0088]

[0089] In formula (9), τ is related to the cutoff frequency of LPF.

[0090] qe′ is a desired sinusoidal signal of 50 Hz, so the cutoff frequency of the LPF can be selected at 50 Hz.

[0091] Step 2.2, establish the improved second-order generalized integrator transfer function, as shown in equation (10) and equation (11), D1(s) and Q1(s) are the two outputs of the improved second-order generalized integrator corresponding to e αβ ',qe αβ '. Compared with the traditional SOGI, the improved SOGI adds a low-pass filter (LPF) to the output error feedback controller kε and qe' channel, and outputs qe' after subtracting it from qe'. The purpose of adding a low-pass filter is to make qe' have a greater attenuation in the high frequency band.

[0092]

[0093]

[0094] Step 2.3, in order to solve the problem of the system operating under the special working condition of unbalanced three-phase power grid, the positive sequence component of the output back electromotive force should be extracted. As shown in formula (12):

[0095]

[0096] The back electromotive force e in equation (9) is α 、e β They are respectively connected to an improved second-order generalized integrator, which outputs two orthogonal back electromotive forces without DC components and high-order harmonics.

[0097] Step 3: Convert the back EMF e α 、e β After the stator flux is input into the integrator to obtain the stator flux, the rotor flux, rotor position and rotor electrical angular velocity can be obtained by equations (2), (4) and (6).

[0098] From the transfer functions D1(s) and Q1(s), it can be analyzed that D1(s) is a second-order bandpass filter, whose bandwidth is only affected by the k value and has nothing to do with the resonant frequency ω'. Outside the center frequency ω', other frequencies will be greatly attenuated. After adding the low-pass filter, the amplitude-frequency characteristics of D1(s) and Q1(s) are still the same, that is, Q1(s) can not only suppress the DC component of the input signal, but also suppress the high-frequency component of the input signal very well, and will not cause delay at the fundamental frequency due to the addition of LPF. Its steady-state gain is also only affected by k and has nothing to do with the resonant frequency ω'. The bandwidth can be set by adjusting the k value. The smaller the k value, the better the filtering effect of the filter, but the longer the dynamic response time.

[0099] For the selection of k value, the filtering effect and response speed of SOGI are taken into consideration. Generally, k is set between Nearby, it also meets the filter quality factor Q D2 =0.707. In order to achieve better dynamic performance, the ω′ in the improved second-order generalized integrator adopts the closed-loop feedback of the estimated linear motor operating frequency.

[0100] The specific steps of step 3 are:

[0101] Step 3.1, the back electromotive force obtained by the second-order generalized integrator is passed through the integrator to obtain the stator flux:

[0102]

[0103] Step 3.2, according to the relationship between the rotor flux and the stator flux, we can get equation (14):

[0104]

[0105] Step 3.3, normalizing the rotor flux can obtain equation (15):

[0106]

[0107] Step 3.4, estimate the position of the mover, the electrical angular velocity of the mover and the speed of the mover by using the obtained mover flux as shown in equations (16)(17)(18):

[0108]

[0109]

[0110]

[0111] The present invention is a method for a sensorless control system of a permanent magnet linear synchronous motor flux observer based on an improved second-order generalized integrator. A feedback channel is added to the improved second-order generalized integrator to enhance the ability to suppress DC components, and a low-pass filter is designed and added to its feedback channel to filter out high-order harmonics caused by end effects and inverter nonlinearity. The extraction of positive sequence components can solve the problem of a special working condition in which the power grid is in a three-phase unbalanced state. The normalization of the rotor flux can eliminate the initial jitter of the speed estimation that may be caused by the initial integration setting in the initial estimation of speed and position.

[0112] The present invention realizes higher dynamic performance and smaller static error of the sensorless control system of the permanent magnet synchronous linear motor, mainly based on the improved second-order generalized integrator to eliminate the flux drift of the flux observer and the higher harmonics caused by the end effect of the linear motor and the nonlinearity of the inverter, thereby improving the observation accuracy of the mover position and speed of the flux observer.

Claims

1. A method for improving sensorless control performance of a permanent magnet synchronous linear motor, characterized in that: The permanent magnet linear synchronous motor flux observer sensorless vector control system based on improved second-order generalized integrator is used for control. The steps are as follows: Step 1, in the α-β coordinate system, a state equation of the permanent magnet synchronous linear motor is obtained, and then the state equation of the permanent magnet synchronous linear motor is described using a state space to obtain the described state equation of the permanent magnet synchronous linear motor; The mathematical model of the permanent magnet synchronous linear motor in step 1 is expressed as: In formula (1), u α 、u β are the components of the stator voltage on the α and β axes respectively; R is the stator resistance; ψ sα , sβ They are the stator flux components of α and β axes respectively; i α 、i β are the components of the stator current on the α and β axes respectively; L s Set as the stator inductance of the permanent magnet linear synchronous motor; The components of the rotor flux on the α and β axes can be expressed as: The relationship between the components of the rotor magnetic flux on the α and β axes and the electrical angle can be expressed as: In formula (3), ψ f is the permanent magnet flux of the permanent magnet linear motor; In formula (4), θ e is the position of the mover The relationship between the mover position and the initial position can be expressed as: In formula (5), θ e is the position of the mover, τ is the pole pitch, and x is the initial position; Therefore, the electrical angular velocity can be expressed as: In formula (6), ω e is the electrical angular velocity, v is the running speed of the linear motor; The calculation of the mover flux is based on the following formula: At this time, the back electromotive force is expressed as follows: Step 2, according to the permanent magnet linear synchronous motor state equation, build an improved second-order generalized integrator model, and define a new back electromotive force output by the second-order generalized integrator; The step 2 comprises: Step 2.1, establish the LPF transfer function formula: qe′ is a 50Hz sinusoidal signal, so the cutoff frequency of LPF can be selected at 50Hz; Step 2.2, the transfer function of the improved second-order generalized integrator is: Where the output of D1(s) is e without DC component αβ ', the output of Q1(s) does not contain DC components and high-order harmonics and is consistent with e αβ 'Orthogonal qe αβ '; For the selection of k value, taking into account the filtering effect and response speed of SOGI, k is selected between Nearby, it also meets the filter quality factor Q D2 =0.707; In order to achieve better dynamic performance, the ω′ in the improved second-order generalized integrator adopts the closed-loop feedback value of the estimated linear motor operating frequency; Step 2.3, the back electromotive force output by the second-order generalized integral must be extracted through the positive sequence component as shown in formula (12): Step 3, in the flux observer, an improved second-order generalized integrator plus an integrator is used to construct an improved flux observer, and a permanent magnet linear synchronous motor flux observer sensorless control system based on the improved second-order generalized integrator is established. The bandwidth gain and frequency parameters of the second-order generalized integrator are further configured by analyzing the transfer function; and the "normalization" method is used to process the rotor flux to solve the jitter problem in the initial estimation of the electrical angular velocity caused by the setting of the initial value of the integral; The specific steps of step 3 are: Step 3.1, the back electromotive force obtained by the second-order generalized integrator is passed through the integrator to obtain the stator flux: Step 3.2, according to the relationship between the rotor flux and the stator flux, we can get equation (14): Step 3.3, normalizing the rotor flux can obtain equation (15): Step 3.4, estimate the position of the mover, the electrical angular velocity of the mover and the speed of the mover by using the obtained mover flux as shown in equations (16)(17)(18):

2. The method for improving the sensorless control performance of a permanent magnet synchronous linear motor according to claim 1, characterized in that: A sensorless vector control system for permanent magnet linear synchronous motor flux observer based on improved second-order generalized integrator d = 0, realizing the dual closed-loop control of current and speed; the three-phase current i a 、i b 、i c After being measured by the Hall sensor, the current i in the two-phase stationary coordinate system is first obtained through a 3s / 2s transformation. α 、i β , and then through 2s / 2r transformation, the two-phase rotating coordinate current i is obtained d 、i q The speed estimate value estimated by the flux observer is subtracted from the given speed value to obtain an error signal, which is then passed through the PI regulator to obtain the q-axis PI regulator current given value, and then the error signal obtained by subtracting it from the actual q-axis current value fed back is converted into the q-axis voltage after passing through the PI regulator. The error signal obtained by subtracting the given d-axis current from the actual value of the d-axis current is converted into the d-axis voltage after passing through the PI regulator. The voltage vector in the dq coordinate system can be obtained, and the voltage vector is then transformed by 2r / 2s to obtain the stationary coordinate system voltage vector u α 、u β Finally, six PWM signals are output through space vector pulse width modulation to drive the motor to move.

Citation Information

Patent Citations

  • Permanent magnet synchronous motor rotor position observation system for restraining position pulsation observation errors and observation method thereof

    CN103560738A

  • Resonant frequency tracking control method for linear oscillation motor

    CN106357183A